Home > Encyclopedia of Public Health > Vital Statistics

Vital statistics are perhaps the most widely used national, state, and local data for identifying and addressing major public health issues. In the United States, legal authority for the registration of vital events (births, deaths, marriages, divorces, fetal deaths, and induced terminations of pregnancy [abortions]) resides with the states, and individually with New York City, the District of Columbia, and the U.S. territories. The states are the legal proprietors of these data and are responsible for maintaining registries and issuing copies of the records.

The existence of a national data system of registration-based vital statistics depends on a cooperative relationship between the states and the federal government. This relationship has evolved over many decades, with its initial beginnings in the early development of the public health movement and the creation of the American federal vital statistics system.

THE HISTORY OF VITAL STATISTICS

The registration of births, marriages, and deaths has a long history in the United States, beginning with registration laws enacted by the Grand Assembly of Virginia in 1632 and the General Court of the Massachusetts Bay Colony in 1639. In enacting this legislation, the early settlers, who were predominantly English, were following English customs. Thus, Virginia law required the clergy to keep a record of all christenings, marriages, and burials in their parishes. The Massachusetts law differed from Virginia's in two important respects: it called for the recording of vital events (births, deaths, and marriages) rather than church-related ceremonies; and it placed responsibility for registration of vital events on government officials rather than the clergy. Little or no statistical use was made of such records, however—along with wills and property inventories, they were regarded primarily as statements of fact essential to the protection of individual rights, especially those relating to the ownership and distribution of property.

The impetus for the use of vital records as the basis of a public health data system came from the realization that records of births and deaths, particularly records of deaths by cause of death, could provide information needed for the control of epidemics and the conservation of human life through sanitary reform. The origin of vital statistics in the modern sense can be traced to an analysis of the English bills of mortality published by John Graunt in 1662. Similarly, the clergyman Cotton Mather noted, in 1721, during a severe smallpox epidemic in Boston, that more than one in six of the natural cases died, but only one in sixty of the inoculated cases died.

In the nineteenth century, the industrial revolution resulted in rapid urbanization, overcrowding of cities, and a deterioration of social and living conditions for large sectors of the population. Public health reformers became acutely conscious of the need for general sanitary reform as a means of controlling epidemics of disease, particularly cholera. These early sanitarians used the crude death statistics of the time to arouse public awareness of the need for improved sanitation, and in the process they pressed for more precise statistics through effective registration practices and laws. The work of Edwin Chadwick (1800–1890) and Dr. William Farr (1807–1883) in England and of Lemuel Shattuck (1793–1859) in Massachusetts was instrumental in the development of public health organization and practice, including the recording of vital statistics. Thus, the history of public health is largely the history of vital registration and statistics.

The United States Constitution provided for a decennial census but not a national vital registration system. To obtain national data on births, marriages, and deaths, the decennial censuses in the latter half of the nineteenth century included questions about vital events, such as: "Born within the year," "Married within the year," and "Disease, if died within the year." These census items were soon recognized as inefficient and the results as deficient. Therefore, when the Bureau of the Census was made a permanent agency of the federal government in 1902, the enabling legislation authorized the bureau to obtain annually copies of records filed in the vital statistics offices of those states and cities having adequate death registration systems and to publish data from these records. This marked the birth of the National Vital Statistics System. Ten states and cities provided death records to the Census Bureau in 1902. In 1915, birth registration was added to the system, and by 1933 all states were registering live births and deaths and providing the required data.

In 1946 responsibility for collecting and publishing national vital statistics was transferred from the Census Bureau to the U.S. Public Health Service, first in the National Office of Vital Statistics and later (1960) in the National Center for Health Statistics (NCHS). In 1987 NCHS became part of the Centers for Disease Control and Prevention of the U.S. Department of Health and Human Services.

In the early part of the twentieth century, the Bureau of the Census received unit record data from the states in hard copy or microfilm. States were reimbursed for copying efforts at four cents per record. Data were transcribed (later key entered) at both the national and state levels, as both states and federal government produced statistics. In 1971 NCHS began an experiment with the state of Florida to receive data on computer tape. This effort expanded and evolved over time, and by 2000, electronic processing and transmission was the norm. NCHS provides partial funding support for state vital statistics efforts and also works with states to implement standards for data elements, editing and coding specifications, quality control procedures, and data transmission schedules.

VITAL STATISTICS DATA FILES

The National Vital Statistics System includes several major electronic data files, each containing the demographic and health information recorded on all events that occur in the United States. Birth data are recorded in the "natality file," which includes characteristics of mother's age, race, Hispanic origin, education, residence, marital status, month of pregnancy, month prenatal care began, tobacco use, and weight gain during pregnancy. Characteristics of the birth include birth weight, length of gestation, sex, plurality, method of delivery, and congenital anomalies.

The "mortality file" variables include residence, place of occurrence, month of death, age, race, Hispanic origin, birthplace, sex, education, marital status, and underlying and multiple causes of death. The "fetal death file" includes data on all fetal deaths of twenty weeks or more gestation. The characteristics of the mother and the delivery are similar to those for natality, but also include the fetal or maternal conditions causing death.

The "linked birth/infant death data system" includes three separate files: a numerator file with linked birth-infant death records for infants who died during the period; a denominator file of data for all births; and a file of the relatively few infant death records that were not linked to birth certificates.

Unlike the natality and mortality systems, detailed data for marriages and divorces have never covered the entire United States. With data year 1996, NCHS ceased collecting detailed marriage and divorce data from the states that had been providing unit records. Monthly counts of the number of marriages and divorces continue to be obtained from each state.

Data and reports from these files are available at the NCHS web site http://www.cdc.gov/nchs/nvss.htm.

VITAL STATISTICS MEASURES

The vital statistics system provides counts of the number of times specified vital events have occurred. These counts are useful in themselves. For example, the numbers of births and deaths are used in the estimation of population size. For most purposes, however, other statistical measures are needed. For example, comparisons of births in one place with those in another requires information on the population size of each area. The simplest and cleanest method of making such comparisons is to compute rates that relate the events to the population exposed to the risk of the event (e.g., the number of births to the number of women of child-bearing age).

Many types of relative numbers are used in the analysis of vital statistics. Those used most frequently in the United States are listed below. The reader can find additional information about the definition and computation of these statistics in the works of Weeks (1996), Shryock and Siegel (1976), or Pollard et al. (1991).

Crude rates. The number of events in a given time period divided by the population at risk produces crude rates. The result is multiplied by a constant (typically 1,000 or 100,000) for ease of presentation. Common crude rates include birth, death, marriage, and divorce.

Specific rates. Crude rates may be limited to a specific group, such as deaths from a specified cause or in a specific age group, or births to unmarried women.

Age-adjusted rates. Age-adjustment is a technique used to eliminate the effect of the age distribution of the population on mortality rates. Since the frequency of death varies with age, a measure free of the influences of population composition is needed to make comparisons between areas or over time.

Infant mortality rates. Infant mortality rates reflect the risk of deaths to infants under the age of one year. For infant deaths, the most commonly used estimate of the population at risk (denominator) is the number of live births during the period.

Life tables and life expectancy. A life table is used to measure the effect of mortality on longevity. It shows the mortality experience of a hypothetical group of infants born at the same time and subject to the mortality rates of a specific population group. A life table provides numerous statistics; perhaps the most widely used is life expectancy at birth.

INTERNATIONAL CLASSIFICATION OF DISEASES

Causes of death are classified for purposes of statistical tabulation according to the International Classification of Diseases (ICD), which is published by the World Health Organization. Traditionally, a single cause of death is selected for statistical tabulations. When the certifying physician indicates that more than one cause contributed to death, a procedure is required for selecting the single cause to be tabulated. The ICD provides the basic ground rules used to code and classify causes of death, to identify the underlying cause of death, and to compensate for inconsistencies in the reported cause-of-death statement. It also includes definitions of terms such as "underlying cause of death," "live birth," "maternal death," as well as tabulation lists which define the cause-of-death groupings to be used for international comparisons. The ICD delineates the format of the medical certification of death and specific regulations regarding the compilation and publication of statistics on diseases and causes of death.

VITAL STATISTICS AND THE PRACTICE OF PUBLIC HEALTH

Over several centuries of development, the vital registration system in the United States has evolved into the primary source of fundamental public health information. Data on deaths, especially causes of death, have been critical for identifying, tracking, and eventually understanding and controlling epidemics of communicable diseases. Today, mortality data are used to study trends and differentials in all kinds of causes of death, both chronic and communicable, as well as those due to homicide, suicide, and unintentional injuries. Infant mortality has traditionally served as a key indicator of general health conditions in a given population. The availability of mortality statistics for small geographic units, such as counties, has contributed uniquely to the value of these data for epidemiologic investigations and surveillance.

Statistics obtained from birth certificates, fetal death reports, and the linked birth/infant death file provide a wealth of information about infant health. Statistics on birth weight, length of gestation, smoking during pregnancy, access to prenatal care, complications of labor and/or delivery, and obstetric procedures are monitored by health care providers and epidemiologists specializing in infant and child health.

In the arena of public policy, vital statistics also provide fundamental information. For example, teen pregnancy and nonmarital childbearing are topics of continuing interest in national welfare policy. Similarly, national health policy is very much concerned with the problem of health disparities among various race and ethnic groups. In these and many other important policy issues the vital statistics system constitutes a frontline source of information that leads to action programs, yields indicators of effectiveness, and generally guides the practice of public health.

Vital statistics are one of the few data systems that are generally available throughout the world. The United Nations and the World Health Organization have led efforts to standardize registration practices, definitions, and statistical measurement. Most countries have at least a rudimentary vital statistics system, and while there are inter-country variations, countries generally adhere to similar registration principles and statistical measures. These data are widely used to make international comparisons of life expectancy, cause-specific mortality, infant deaths, and other important measures. Vital statistics are also used for monitoring population growth, through measures such as total fertility rates. The United Nations publishes many international vital statistics comparisons in its Demographic Yearbook.

MARY ANNE FREEDMAN

JAMES A. WEED

(SEE ALSO: Abortion; Bills of Mortality; Biostatistics; Birth Certificates; Birthrate; Census; Certification of Causes of Death; Chadwick, Edwin; Farr, William; Graunt, John; Infant Mortality Rate; International Classification of Diseases; Life Expectancy and Life Tables; Mortality Rates; National Center for Health Statistics; Rates; Rates: Adjusted; Rates: Age-Adjusted; Shattuck, Lemuel; Statistics for Public Health)